Search for: All records

Task‐incremental learning (Task‐IL) aims to enable an intelligent agent to continuously accumulate knowledge from new learning tasks without catastrophically forgetting what it has learned in the past. It has drawn increasing attention in recent years, with many algorithms being proposed to mitigate neural network forgetting. However, none of the existing strategies is able to completely eliminate the issues. Moreover, explaining and fully understanding what knowledge and how it is being forgotten during the incremental learning process still remains under‐explored. In this paper, we propose KnowledgeDrift, a visual analytics framework, to interpret the network forgetting with three objectives: (1) to identify when the network fails to memorize the past knowledge, (2) to visualize what information has been forgotten, and (3) to diagnose how knowledge attained in the new model interferes with the one learned in the past. Our analytical framework first identifies the occurrence of forgetting by tracking the task performance under the incremental learning process and then provides in‐depth inspections of drifted information via various levels of data granularity. KnowledgeDrift allows analysts and model developers to enhance their understanding of network forgetting and compare the performance of different incremental learning algorithms. Three case studies are conducted in the paper to further provide insights and guidance for users to effectively diagnose catastrophic forgetting over time.

 

Transcriptome studies that provide temporal information about transcript abundance facilitate identification of gene regulatory networks (GRNs). Inferring GRNs from time series data using computational modeling remains a central challenge in systems biology. Commonly employed clustering algorithms identify modules of like-responding genes but do not provide information on how these modules are interconnected. These methods also require users to specify parameters such as cluster number and size, adding complexity to the analysis. To address these challenges, we used a recently developed algorithm, partitioned local depth (PaLD), to generate cohesive networks for 4 time series transcriptome datasets (3 hormone and 1 abiotic stress dataset) from the model plant Arabidopsis thaliana. PaLD provided a cohesive network representation of the data, revealing networks with distinct structures and varying numbers of connections between transcripts. We utilized the networks to make predictions about GRNs by examining local neighborhoods of transcripts with highly similar temporal responses. We also partitioned the networks into groups of like-responding transcripts and identified enriched functional and regulatory features in them. Comparison of groups to clusters generated by commonly used approaches indicated that these methods identified modules of transcripts that have similar temporal and biological features, but also identified unique groups, suggesting that a PaLD-based approach (supplemented with a community detection algorithm) can complement existing methods. These results revealed that PaLD could sort like-responding transcripts into biologically meaningful neighborhoods and groups while requiring minimal user input and producing cohesive network structure, offering an additional tool to the systems biology community to predict GRNs.

 

Nitrogen doping in carbon materials can modify the employed carbon material’s electronic and structural properties, which helps in creating a stronger metal-support interaction. In this study, the role of nitrogen doping in improving the durability of Pt catalysts supported on a three-dimensional vertically aligned carbon nanofiber (VACNF) array towards oxygen reduction reaction (ORR) was explored. The nitrogen moieties present in the N-VACNF enhanced the metal-support interaction and contributed to a reduction in the Pt particle size from 3.1 nm to 2.3 nm. The Pt/N-VACNF catalyst showed better durability when compared to Pt/VACNF and Pt/C catalysts with similar Pt loading. DFT calculations validated the increase in the durability of the Pt NPs with an increase in pyridinic N and corroborated the molecular ORR pathway for Pt/N-VACNF. Moreover, the Pt/N-VACNF catalyst was found to have excellent tolerance towards methanol crossover. 

Vertically aligned carbon nanofibers (VACNFs) are promising supports for oxygen reduction reaction (ORR) electrocatalysts in fuel cells. Although experimentally these catalytic systems have shown great potential, there is lack of molecular understanding of the catalytic sites and reaction mechanisms. This work investigated the origin of the ORR reactivities of the platinum catalysts on multi‐edged VACNFs (Pt/VACNF) using a multiscale modeling approach combining Density Functional Theory (DFT) and classical Molecular Dynamics (MD) simulations. Based on the ReaxFF potential, all nanoscale Pt particles (Pt₅₅, P₁₀₀, and Pt₁₄₇) are stabilized by the open edges located axially along the VACNF walls. The calculated first‐shell coordination numbers,, of surface Pt atoms are 6.63, 7.27, and 7.85, respectively, suggesting that the percentage of low coordination sites increases as the particle size decreases. The adsorption energies of OOH, O, and OH on Pt₅₅were systematically probed using DFT calculations. These adsorption energies retain a linear correlation against the generalized coordination numbers (). For Pt nanoparticles supported on VACNF, we found that the OOH and OH bind stronger than on Pt (111) by 0.14 and 0.17 eV, respectively, which can hinder the ORR activity with lower limiting potential than Pt (111). Our theoretical prediction is in good agreement with the linear sweeping voltammetry that revealed a left shift of the half‐wave potential.